PCWrEooUrckResehdoinpgs ISSNc1e6u1r-3w-0s0.o7r3g
i
The sixth edition (PatentSemTech2025) of the workshop series Patent Text Mining and Semantic Technologies was held as a full-day event in conjunction with the SIGIR 2025 conference. As in the previous editions, the workshop focused on new developments and research in patent retrieval and patent analytics. An important focus of the workshop was to address the adaptation of existing deep learning models, e.g. large language models, for the patent domain, covering diverse scientific subject areas, such as chemistry, pharmacology, etc. In general, patent data is more dificult to analyze compared to corpora comprising other text genres. Working with patent data, in addition to its challenging aspects, does bring a richness of facets to be exploited with text mining and semantic analysis methods as well: (1) It constitutes a huge corpus of scientific-technical documents for a variety of technological domains. (2) They are rich in available meta-data such as spatial data, bibliographic data, classifications, temporal data, etc. (3) Patents describe essential scientific-technical knowledge that encloses solutions for real-world applications. (4) They are complementary knowledge to scientific literature, mostly not published elsewhere.
With the PatentSemTech2025 workshop, we continue our series of workshops launched in 2019, aiming to establish long-term collaboration and a two-way communication channel between the IP industry and academia from relevant fields. Therefore, the 6th PatentSemTech workshop was organized as a full-day event with 9 presentations of research papers that were accepted after peer review out of 19 submissions. We had 4 long paper presentations and 5 short paper presentations; the latter were also presented in the joint workshops’ poster presentation. In addition, Iryna Gurevych (TU Darmstadt, Germany) gave a keynote speech on "How to InterText? Elevating NLP to the cross-document level". In an expert panel moderated by Alexander Klenner (European Patent Ofice, The Netherlands) six experts discussed how generative AI is changing the patent domain and experiences were exchanged with the workshop participants on how to apply AI for patent-related tasks.
Hidir Aras, Linda Andersson,
Florina Piroi, Allan Hanbury, Dean Alderucci Organizers Program Committee
More information on the topics, schedule, and further developments of the PatentSemTech workshop can be found on the website: http://ifs.tuwien.ac.at/patentsemtech/